similar to: Estimate Fisher Information by Hessian from OPTIM

Special Volume on Spectroscopy and Chemometrics in R Guest Edited by Katharine M. Mullen and Ivo H.M. van Stokkum 11 contributions, with software packages and examples Similar volumes on R in Psychometrics (guest edited by Jan de Leeuw) and R in Political Methodology (guest edited by Micah Altman and Simon Jackman) are close to being finished. Additional volumes are being worked on. Proposals

Dear helpers, I am using the function "constrOptim" to estimate a model with ML with an inequality constraint using the option method='Nelder-Mead'. When I specify the option: hessian = TRUE I obtain the response: Error in f(theta, ...) : unused argument(s) (hessian = TRUE) I guess the function "constrOptim" does not allow this argument which, on the other hand, is

Dear Helper, I used "constrOptim.nl" and got the value of par. The estimations looks good even if the number of iterations is only 16. But the values of hessian and convergence are both "NULL". I tested the objective function and gradient function by "optim" and didn't see any problem there. With these functions, "optim" gives the convergence value

Dear R-users, I'm estimating a two- dimensional state-space model using the FKF package. The resulting log likelihood function is maximized using auglag from the Alabama package. The procedure works well for a subset of my data, but if I try to use the entire data set I get the following error message. Error in eigen(a$hessian) : infinite or missing values in 'x' What's even

Hello R users, I am trying to get the Std. errors of some estimators using the optim function. But when computing them using sqrt(diag(solve(ans$hessian))) I get an error that says... Lapack routine dgesv: system is exactly singular. Or sometimes I also get that system is computationally singular. When I check the Hessian matrix it has at least one column&row of mostly zeros. My

Hi, Use the `auglag' function in "alabama" if you want to get the Hessian at convergence. This typically tends to perform better than `constrOptim.nl'. Also, `constrOptim.nl' does not compute the Hessian. You should not specify method="L-BFGS-B". The default method "BFGS" is better in this setting. Hope this helps, Ravi

Dear all, I am tryng to solve a nonlinear optimization probel using the solnp function. I have different datasets. For the smaller I get full solutions, for the bigger I got an error message stating: ######################################## Iter: 1 fn: 101.8017 Pars: 0.21000 0.21000 0.21000 0.21000 0.21000 0.21000 0.21000 0.21000 0.21000 0.21000 0.21000 0.21000 0.21000 0.21000 0.21000

Hi, I am working or arima. I think arima uses non-linear optimisation for parameter optimisation. The standard error for parameters are computed from hessian matrix. When I use arima model, how can I see the finial hessian got from non-linear optimisation (BFGS for example). Any help is appreciated. Many thanks. Di [[alternative HTML version deleted]]

Out of curiosity, I have tried, without success, to use the new facility in nlm to specify the gradient and hessian. (It is many years since I had a problem simple enough to make analytic derivation of these worthwhile.) The help now says that the function must have attributes with these names but gives no indication as to what should be in the attributes. The online example and demo do not use

Hi, Is there any way to get the hessian matrix from the "constrOptim" function without supplying gradient function? Thanks. --------------------------------- Bring words and photos together (easily) with [[alternative HTML version deleted]]

Dear all, I am trying to minimize a function with 3 parameters using nlm. I have worked out the 2nd derivatives (incl the cross-product terms) and would like to supply them to nlm for evaluation. What I am not sure about is how to set up the Hessian matrix for nlm. That is, attr(lhat, "hessian") <- c(???) Do I have to enter all 9 of the entries or just the lower triangle of the

Hello! Looking on how people use optim to get MLE I also noticed that one can use returned Hessian to get corresponding standard errors i.e. something like result <- optim(<< snip >>, hessian=T) result$par # point estimates vc <- solve(result$hessian) # var-cov matrix se <- sqrt(diag(vc)) # standard errors What is actually Hessian representing here?

Dear R-users, In the function constrOptim there is an option to get an approximation to the hessian of the surrogate function R at MLE by declaring hessian=TRUE in the calls to the function optim. I would like to ask if it is advisable to get an approximate hessian for the funcrion f as follows: f''(theta)=R''(theta|theta_k)-B''(theta) where

Hi R-community, I've been trying to fit a model using maximum likelihood in nlm. Upon convergence I examine the hessian and it has a few (not all) positive values on its diagonal, as does its inverse. I am minimizing the negative log likelihood, so I would expect all the members of the diagonal to be negative (I think). Can anyone shed light on how this might be interpreted? Thanks much,

Dear R experts, In R-intro, under the 'Nonlinear least squares and maximum likelihood models' there are ttwo examples considered how to use 'nlm' function. In 'Least squares' the Standard Errors obtained as follows: After the fitting, out$minimum is the SSE, and out$estimates are the least squares estimates of the parameters. To obtain the approximate standard

Hi, i have made a code for optimizing a function using "constrOptim". i need hessain matrix of the parameters. how could i get hessain matrix when i will use "constrOptim"? May i get get any help from anyone? thank you in advance. Kanak Choudhury. [[alternative HTML version deleted]]

I am new to 'R' and also new to the concept of a 'Hessian' with non-linear optimization. I would like to avoid going through all of the reference articles given with ?optim as access to a library is not handy. Would someone be able to elighten me on what is in the Hessian matrix if 'hessian = TRUE' with optim? The documentation indicates that this is a numerically

Dear R experts! I try to minimize a function with external C fitting function. I get the hessian matrix. Here it is: [,1] [,2] [,3] [,4] [1,] 1.8816631 0 0.8859803 0 [2,] 0.0000000 0 0.0000000 0 [3,] 0.8859803 0 0.4859983 0 [4,] 0.0000000 0 0.0000000 0 Second and fourth rows/columns have zero values only. That's OK, because that ones related

Hi, R developers. when running mle inside a loop I found a nasty behavior. From time to time, my model had a degenerate minimum and the loop just crashed. I tracked it down to "vcov <- if (length(coef)) solve(oout$hessian)" line, being the hessian singular. Note that the minimum reached was good, it just did not make sense to calculate the covariance matrix as the inverse of a

Hi there! I´m a noob when it comes to R and I´m using it to run statisc analysis. With the code for ARIMA below I´m getting this error: Error in solve.default(res$hessian * n.used) :Lapack routine dgesv: system is exactly singular The code is: > s.ts <- ts(x[,7], start = 2004, fre=12) > get.best.arima <- function (x.ts, maxord=c(1,1,1,1,1,1)) + { + best.aic <- 1e8 + n <-